A Few (More) Words on Commodities

A year or so after initiating a position in commodities, we thought we should look back and evaluate our experience.  To quote Dickens: it was the best of times, it was the worst of times.  After a very strong 2nd quarter for commodities (during a weak market for common stocks equities), commodities were very weak in the 3rd quarter (in a very strong equity market) and weak in the 4th (in an even stronger equity market).

While we never enjoy seeing losses, commodities, with their very low (and, recently, negative) correlation with the equity markets play a very important role in diversifying Kochis Fitz client portfolios.  Indeed, some of the very forces that are driving high recent equity returns (e.g., lower oil and gasoline prices) are directly responsible for the recent negative performance in commodities.  We continue to expect that commodities' long-term returns are likely to be attractive (our very long-term forecast is 8.5%), but importantly, and by design, we'll often get those returns at times other than when the rest of the portfolio is doing well. 

A study of commodities performance for the period 1959 through 2004 published earlier this year found that during the weakest 5 percent of monthly equity markets performance--when stocks fell by 8.98 percent, on average, a month--commodity futures experienced a positive average return of 1.03 percent during those months.  This is slightly above the full sample’s average return of 0.89 percent a month.  During the weakest 1 percent of monthly equity markets performance--when equities fell, on average, by 13.87 percent a month--commodity futures returned an average of 2.38 percent a month.

From a portfolio construction standpoint, a key benefit is that adding some commodities exposure to a portfolio helps offset some of the downside risk in equities.  This permits clients to be more aggressive with their equity exposure otherwise, thus either keeping total expected risk constant for greater return, or reducing risk while keeping expected return constant.  The hard part is being patient with commodities when equity market returns are stellar.

The Kochis Ftiz Commodities Implementation

After substantial research, we chose the Goldman Sachs Commodity Index (GSCI) as our preferred benchmark.  We prefer this to competing commodities indexes which are less volatile on a stand-alone basis but which do not offer the same level of portfolio diversification.  The GSCI reflects the return to holding futures contracts on a basket of commodities whose weights are determined based on worldwide production.

Having chosen GSCI as our benchmark, we then searched for cost-effective and tax-efficient vehicles to deliver the desired exposure.  The Oppenheimer Commodities Strategy Total Return Fund, which was the only attractively priced fund benchmarked to the GSCI, was our first pick.  When it closed to new investors in April 2006, we chose the PIMCO Commodity Real Return Fund, even though not benchmarked to GSCI (it mimics the Dow Jones AIG Commodity Index (DJAIGCI)), for new commodity investments.  Recently, Barclay's Global Investors issued an exchange traded note benchmarked to the GSCI under the name “iPath” (ticker: GSP).  The note offers a lower cost and is more tax efficient than the other funds.   The tax efficiency of the iPath note makes GSP an ideal holding for taxable accounts.  We have not yet fully replaced Oppenheimer or PIMCO with iPath in all client accounts because of asset location and tax considerations, but will continue to look for opportunities to do so in the coming year.

So how are the funds doing?  The Oppenheimer fund has done very well, offering a positive spread to the GSCI, and with less volatility.  And yet its correlation with the GSCI is close to one, indicating that the portfolio diversification benefits have remained.  The PIMCO fund has modestly underperformed its benchmark, though its absolute performance has been better than Oppenheimer and GSCI. 

Interpreting Comparative Data

For those clients who ask about the performance of our choice of commodities exposure relative to supposed alternatives, we must caution that it is very easy to be confused (and mis-led).  For example, Morningstar classifies commodities focused funds, including the Oppenheimer and PIMCO funds, in the “natural resources” category.  This is a catch-all category for a broad range of funds with holdings (primarily equities) loosely defined as natural resources and commodity-related.  The appropriate index for most of the funds in this category is the Goldman Sachs Natural Resources Index (GSNRI), which measures the performance of companies which participate (explore, develop, transport, eg, Exxon) in natural resources rather than the commodities (oil, for example) themselves. 

However, this Natural Resources Index is not an appropriate benchmark for our commodities investments.  Our investments benchmark to either the GSCI or DJAIGCI, and invest primarily in futures contracts and structured notes for various commodities.  As the correlation matrix below shows, Oppenheimer had a correlation of 1.0 with the GSCI, while PIMCO had a correlation of 0.93 with the DJAIGCI.  Over this same period, the correlation of the GSCI and the DJAIGCI with the GSNRI was 0.48 and 0.44, respectively.  Meanwhile, the average natural resources fund had a correlation of 0.98 with the GSNRI.   (As a reminder, a correlation of 1 means that the returns of two funds are strongly positively related, a correlation of 0 means the returns of two funds are not related, and a correlation of -1 means the items are strongly negatively related).  Comparing our commodity investments and the GSCI or DJAIGCI (apples) to GSNRI and a group of natural resource funds (oranges), is inappropriate.

Next Steps

All of this leads us to say that we remain very comfortable with clients’ current allocation to commodities, despite our regret that many clients have actually lost money in 2006 in this asset class.  We also are happy with Oppenheimer, PIMCO, and iPath, broadly, as the implementation opportunities for these allocations. 

This is not to say that our work on commodities is complete.  Recent academic research suggests that commodities returns correlate with some of the same dimensions of risk exposure as equity returns.  For example, “value” (defined a bit differently than in the equity space) is an important predictor of commodity futures returns.  While we overweight value sectors in equity asset allocation and use funds specifically constructed to help us capture the value premium in equities, there are not yet any vehicles which would allow us to capture this value premium in commodities.  But we believe they can be developed and expect to find one … or perhaps develop one ourselves.  This investment category is dynamic, and we are committed to remaining on the forefront of developments to optimize our clients’ exposure to it.

Karen Blodgett, Monica Ma, and Jason Thomas